GPS:graph contrastive learning via multi-scale augmented views from adversarial pooling  

作　　者：Wei JU Yiyang GU Zhengyang MAO Ziyue QIAO Yifang QIN Xiao LUO Hui XIONG Ming ZHANG 

机构地区：[1]School of Computer Science,National Key Laboratory for Multimedia Information Processing,Peking University,Beijing 100871,China [2]Artificial Intelligence Thrust,The Hong Kong University of Science and Technology,Guangzhou 511453,China [3]Department of Computer Science,University of California,Los Angeles CA 90095,USA

出　　处：《Science China(Information Sciences)》2025年第1期142-155,共14页中国科学(信息科学)(英文版)

基　　金：supported by National Natural Science Foundation of China(Grant Nos.62306014,62276002);China Postdoctoral Science Foundation(Grant No.2023M730057).

摘　　要：Self-supervised graph representation learning has recently shown considerable promise in a range of fields,including bioinformatics and social networks.A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs,which train models by maximizing agreement between original graphs and their augmented views(i.e.,positive views).Unfortunately,these methods usually involve pre-defined augmentation strategies based on the knowledge of human experts.Moreover,these strategies may fail to generate challenging positive views to provide sufficient supervision signals.In this paper,we present a novel approach named graph pooling contrast(GPS)to address these issues.Motivated by the fact that graph pooling can adaptively coarsen the graph with the removal of redundancy,we rethink graph pooling and leverage it to automatically generate multi-scale positive views with varying emphasis on providing challenging positives and preserving semantics,i.e.,strongly-augmented view and weakly-augmented view.Then,we incorporate both views into a joint contrastive learning framework with similarity learning and consistency learning,where our pooling module is adversarially trained with respect to the encoder for adversarial robustness.Experiments on twelve datasets on both graph classification and transfer learning tasks verify the superiority of the proposed method over its counterparts.

关 键 词：graph representation learning graph neural networks graph contrastive learning graph augmentations graph pooling 

分 类 号：P228.4[天文地球—大地测量学与测量工程]